DeFP

A tool reading ground truth data and detected object data provides visualization way to estimate position accuracy.

Safety regions research is a well-known task for ML and the main focus is to avoid false positives, i.e., including in the safe region unsafe points. In this repository, two methods for the research of zero FPR regions are proposed: the first one is based simply on the reduction of the SVDD radius until only safe points are enclosed in the SVDD …

Simple code giving the probability from the true versus the false positive rate

Approximate palindromes lead to overstated false match rates in benchmarks using reversed sequences

Machine learning for credit card default. Precision-recalls are calculated due to imbalanced data. Confusion matrices and test statistics are compared with each other based on Logit over and under-sampling methods, decision tree, SVM, ensemble learning using Random Forest, Ada Boost and Gradient Boosting. Easy Ensemble AdaBoost classifier appear…

👞 SequenceSense | A Robust Gesture Sequencing Tool

Slides of a talk on deep learning, false negatives/positives and predator-prey interactions with R.

Group Project, CS454, AI Based Software Engineering, 2019 Fall

Notícia Chique (NC) é Um projeto de site feito em grupo pode envolver diversos desafios e aprendizados valiosos para os envolvidos. No caso do site de notícias sobre "fake news".

Numerical methods programs solution.

"Publication bias and the canonization of false facts" published in eLife (2016)

False positif sample submit

Proof of concept of a false-positive-based anti SYN scan

RMarkdown file showing an example of using the Random Forest algorithm to examine Weather Data

Interactive Tool for Interpreting positive COVID-19 antibody tests

Machine learning model to predict churn of credit card customers. Demonstration of novel XGBoost configuration for noisy, imbalanced data.

This script it's used to disable the attribute 'to_ids' on MISP events, features removal of the IDS tag on old events or based on Reputational datasets results.